Auditory Processing and Reading Disability: A Systematic Review and Meta-Analysis

ABSTRACT

Purpose

Reading disability (RD) is frequently associated with deficits in auditory processing (i.e., processing speech and non-linguistic sounds). Several hypotheses exist regarding the link between RD and auditory processing, but none fully account for the range/variety of auditory impairments reported in the literature. These impairments have been primarily summarized by qualitative reviews and meta-analytic evidence for most auditory processing impairments is lacking.

Method

We conducted a PRISMA-compliant meta-analysis quantifying the degree to which individuals with RD are impaired on four categories of auditory processing abilities: frequency discrimination, intensity discrimination, duration discrimination, and gap detection. This methodology was accepted and executed as a Registered Report.

Results

Auditory processing impairments of medium to large effect size were present in RD vs. typical groups for all categories: frequency (g = 0.79), duration (g = 0.80), and intensity discrimination (g = 0.60), as well as gap detection (g = 0.80). No differences were found across task designs (i.e., testing methods).

Conclusion

This meta-analysis documents a large, multiple-domain non-linguistic, auditory processing impairment in RD. Contrary to previous studies, we found a significant deficit in intensity discrimination. The impairments described here must be accounted for by future causal hypotheses in RD and suggest that auditory processing impairments are broader than previously thought.

Acknowledgments

We thank Rick Qian, Gabriella Leibowitz, and Tessneem Shahbandar for their assistance with data collection and extraction. We also thank the many authors who sent us their data to be included in this meta-analysis. We thank Beth Tipton for essential statistical guidance. We thank Sumit Dhar and Megan Roberts for input and comments on earlier versions of this manuscript. The authors declare that they have no conflict of interest.

Disclosure statement

No potential conflict of interest was reported by the authors.

Data sharing and accessibility

We have shared all raw data, digital materials, and analysis code, as well as registered our approved protocol on the study’s Open Science Framework page (https://osf.io/nwctx).

Ethical approval

As this study is a systematic review and meta-analysis, ethical approval is not required.

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/10888438.2023.2252118.

Notes

1. Here, we use the term RD to describe a primary deficit in reading accuracy, speed, or comprehension; this is a slightly broader term than developmental dyslexia, which is typically defined by deficits primarily at the word level, though the terms are sometimes used interchangeably (Peterson & Pennington, 2015).

2. There is considerable debate in the field about the criteria for APD (see Chermak et al., 2018; Vermiglio, 2018). This is orthogonal to the current analyses because isolated, continuous auditory task scores are used, thus this debate will not be discussed here.

3. The Stage 1 submission of this manuscript contained the wording “children with RD” rather than individuals with RD in the methods, when the intention was to include all ages. This oversight was presented plainly to the editors and reviewers.

4. The study quality tool was adapted from its original form in order to better fit the study designs and constructs present in the literature. The original tool and the modified tool are presented in Supplemental Materials.

5. During final code review before publishing, authors found that power estimate calculation should have been ≥.99 rather than the value of .96 reported in the Stage 1 Registered Report.

6. The sample sizes (n)s of each age group do not add to the full sample because Goswami et al. (2010) and Thomson and Goswami (2008) are part of the same longitudinal study, in which the subjects were children at the initial timepoint and are adolescents by the final reported timepoint. These participants are counted twice only in descriptive statistics and are considered correlated effects in the meta-analyses.

7. Due to differences in the format of a dissertation vs. a published journal article with the same data (Zaidan, 2009; Zaidan & Baran, 2013, respectively), one study had ratings as both “poor” and “fair.” The “fair” study quality rating is used here.

Additional information

Funding

This work was funded by Northwestern University including the Graduate Research Grant and Office of Undergraduate Research. Research reported in this publication was supported, in part, by the National Institutes of Health's National Center for Advancing Translational Sciences, Grant Number UL1TR001422. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.